Hugging Face's logo

hmbert-tiny
/
flair-hipe-2022-letemps-fr

Token Classification
Flair
PyTorch
TensorBoard
French
sequence-tagger-model
Model card Files Metrics Community
stefan-it commited on
Commit
6685b53
•
1 Parent(s): e25e301

Upload folder using huggingface_hub

Browse files
Files changed (6) hide show
  1. best-model.pt +3 -0
  2. dev.tsv +0 -0
  3. loss.tsv +11 -0
  4. runs/events.out.tfevents.1697671916.46dc0c540dd0.3802.1 +3 -0
  5. test.tsv +0 -0
  6. training.log +245 -0
best-model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f7602a47e24b949faf5f72aab9040dd58eb26c8e298809b00ce78f28cbdef6a
3
+ size 19045922
dev.tsv ADDED
The diff for this file is too large to render. See raw diff
 
loss.tsv ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ EPOCH TIMESTAMP LEARNING_RATE TRAIN_LOSS DEV_LOSS DEV_PRECISION DEV_RECALL DEV_F1 DEV_ACCURACY
2
+ 1 23:32:55 0.0000 0.6062 0.1752 0.2635 0.3021 0.2814 0.1683
3
+ 2 23:33:55 0.0000 0.1692 0.1676 0.3814 0.4027 0.3918 0.2493
4
+ 3 23:34:55 0.0000 0.1415 0.1629 0.4301 0.5355 0.4771 0.3225
5
+ 4 23:35:56 0.0000 0.1259 0.1716 0.4394 0.5595 0.4922 0.3365
6
+ 5 23:36:55 0.0000 0.1122 0.1804 0.4415 0.5915 0.5056 0.3481
7
+ 6 23:37:56 0.0000 0.1017 0.1968 0.4447 0.6259 0.5200 0.3594
8
+ 7 23:38:57 0.0000 0.0964 0.2049 0.4573 0.5824 0.5123 0.3520
9
+ 8 23:39:58 0.0000 0.0898 0.2238 0.4519 0.6293 0.5261 0.3657
10
+ 9 23:40:58 0.0000 0.0858 0.2332 0.4555 0.6327 0.5297 0.3684
11
+ 10 23:41:59 0.0000 0.0830 0.2346 0.4558 0.6259 0.5275 0.3671
runs/events.out.tfevents.1697671916.46dc0c540dd0.3802.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ae9cf88eddc42bf2a2f42e1f651911d19dd31ec124a4dbb2ecaad46ecef5df1
3
+ size 2030580
test.tsv ADDED
The diff for this file is too large to render. See raw diff
 
training.log ADDED
@@ -0,0 +1,245 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ 2023-10-18 23:31:56,783 ----------------------------------------------------------------------------------------------------
2
+ 2023-10-18 23:31:56,783 Model: "SequenceTagger(
3
+ (embeddings): TransformerWordEmbeddings(
4
+ (model): BertModel(
5
+ (embeddings): BertEmbeddings(
6
+ (word_embeddings): Embedding(32001, 128)
7
+ (position_embeddings): Embedding(512, 128)
8
+ (token_type_embeddings): Embedding(2, 128)
9
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
10
+ (dropout): Dropout(p=0.1, inplace=False)
11
+ )
12
+ (encoder): BertEncoder(
13
+ (layer): ModuleList(
14
+ (0-1): 2 x BertLayer(
15
+ (attention): BertAttention(
16
+ (self): BertSelfAttention(
17
+ (query): Linear(in_features=128, out_features=128, bias=True)
18
+ (key): Linear(in_features=128, out_features=128, bias=True)
19
+ (value): Linear(in_features=128, out_features=128, bias=True)
20
+ (dropout): Dropout(p=0.1, inplace=False)
21
+ )
22
+ (output): BertSelfOutput(
23
+ (dense): Linear(in_features=128, out_features=128, bias=True)
24
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
25
+ (dropout): Dropout(p=0.1, inplace=False)
26
+ )
27
+ )
28
+ (intermediate): BertIntermediate(
29
+ (dense): Linear(in_features=128, out_features=512, bias=True)
30
+ (intermediate_act_fn): GELUActivation()
31
+ )
32
+ (output): BertOutput(
33
+ (dense): Linear(in_features=512, out_features=128, bias=True)
34
+ (LayerNorm): LayerNorm((128,), eps=1e-12, elementwise_affine=True)
35
+ (dropout): Dropout(p=0.1, inplace=False)
36
+ )
37
+ )
38
+ )
39
+ )
40
+ (pooler): BertPooler(
41
+ (dense): Linear(in_features=128, out_features=128, bias=True)
42
+ (activation): Tanh()
43
+ )
44
+ )
45
+ )
46
+ (locked_dropout): LockedDropout(p=0.5)
47
+ (linear): Linear(in_features=128, out_features=13, bias=True)
48
+ (loss_function): CrossEntropyLoss()
49
+ )"
50
+ 2023-10-18 23:31:56,783 ----------------------------------------------------------------------------------------------------
51
+ 2023-10-18 23:31:56,783 MultiCorpus: 14465 train + 1392 dev + 2432 test sentences
52
+ - NER_HIPE_2022 Corpus: 14465 train + 1392 dev + 2432 test sentences - /root/.flair/datasets/ner_hipe_2022/v2.1/letemps/fr/with_doc_seperator
53
+ 2023-10-18 23:31:56,783 ----------------------------------------------------------------------------------------------------
54
+ 2023-10-18 23:31:56,783 Train: 14465 sentences
55
+ 2023-10-18 23:31:56,783 (train_with_dev=False, train_with_test=False)
56
+ 2023-10-18 23:31:56,783 ----------------------------------------------------------------------------------------------------
57
+ 2023-10-18 23:31:56,783 Training Params:
58
+ 2023-10-18 23:31:56,783 - learning_rate: "5e-05"
59
+ 2023-10-18 23:31:56,783 - mini_batch_size: "4"
60
+ 2023-10-18 23:31:56,783 - max_epochs: "10"
61
+ 2023-10-18 23:31:56,783 - shuffle: "True"
62
+ 2023-10-18 23:31:56,783 ----------------------------------------------------------------------------------------------------
63
+ 2023-10-18 23:31:56,783 Plugins:
64
+ 2023-10-18 23:31:56,783 - TensorboardLogger
65
+ 2023-10-18 23:31:56,783 - LinearScheduler | warmup_fraction: '0.1'
66
+ 2023-10-18 23:31:56,783 ----------------------------------------------------------------------------------------------------
67
+ 2023-10-18 23:31:56,783 Final evaluation on model from best epoch (best-model.pt)
68
+ 2023-10-18 23:31:56,784 - metric: "('micro avg', 'f1-score')"
69
+ 2023-10-18 23:31:56,784 ----------------------------------------------------------------------------------------------------
70
+ 2023-10-18 23:31:56,784 Computation:
71
+ 2023-10-18 23:31:56,784 - compute on device: cuda:0
72
+ 2023-10-18 23:31:56,784 - embedding storage: none
73
+ 2023-10-18 23:31:56,784 ----------------------------------------------------------------------------------------------------
74
+ 2023-10-18 23:31:56,784 Model training base path: "hmbench-letemps/fr-dbmdz/bert-tiny-historic-multilingual-cased-bs4-wsFalse-e10-lr5e-05-poolingfirst-layers-1-crfFalse-1"
75
+ 2023-10-18 23:31:56,784 ----------------------------------------------------------------------------------------------------
76
+ 2023-10-18 23:31:56,784 ----------------------------------------------------------------------------------------------------
77
+ 2023-10-18 23:31:56,784 Logging anything other than scalars to TensorBoard is currently not supported.
78
+ 2023-10-18 23:32:02,616 epoch 1 - iter 361/3617 - loss 2.90306109 - time (sec): 5.83 - samples/sec: 6477.10 - lr: 0.000005 - momentum: 0.000000
79
+ 2023-10-18 23:32:08,209 epoch 1 - iter 722/3617 - loss 1.99373894 - time (sec): 11.43 - samples/sec: 6675.78 - lr: 0.000010 - momentum: 0.000000
80
+ 2023-10-18 23:32:13,920 epoch 1 - iter 1083/3617 - loss 1.43612253 - time (sec): 17.14 - samples/sec: 6767.81 - lr: 0.000015 - momentum: 0.000000
81
+ 2023-10-18 23:32:19,644 epoch 1 - iter 1444/3617 - loss 1.15677364 - time (sec): 22.86 - samples/sec: 6745.40 - lr: 0.000020 - momentum: 0.000000
82
+ 2023-10-18 23:32:24,914 epoch 1 - iter 1805/3617 - loss 0.98282051 - time (sec): 28.13 - samples/sec: 6867.21 - lr: 0.000025 - momentum: 0.000000
83
+ 2023-10-18 23:32:30,238 epoch 1 - iter 2166/3617 - loss 0.86833073 - time (sec): 33.45 - samples/sec: 6872.25 - lr: 0.000030 - momentum: 0.000000
84
+ 2023-10-18 23:32:35,906 epoch 1 - iter 2527/3617 - loss 0.77953327 - time (sec): 39.12 - samples/sec: 6811.87 - lr: 0.000035 - momentum: 0.000000
85
+ 2023-10-18 23:32:41,535 epoch 1 - iter 2888/3617 - loss 0.71123326 - time (sec): 44.75 - samples/sec: 6776.65 - lr: 0.000040 - momentum: 0.000000
86
+ 2023-10-18 23:32:47,198 epoch 1 - iter 3249/3617 - loss 0.65498355 - time (sec): 50.41 - samples/sec: 6744.88 - lr: 0.000045 - momentum: 0.000000
87
+ 2023-10-18 23:32:53,052 epoch 1 - iter 3610/3617 - loss 0.60698836 - time (sec): 56.27 - samples/sec: 6741.68 - lr: 0.000050 - momentum: 0.000000
88
+ 2023-10-18 23:32:53,154 ----------------------------------------------------------------------------------------------------
89
+ 2023-10-18 23:32:53,154 EPOCH 1 done: loss 0.6062 - lr: 0.000050
90
+ 2023-10-18 23:32:55,378 DEV : loss 0.17521269619464874 - f1-score (micro avg) 0.2814
91
+ 2023-10-18 23:32:55,404 saving best model
92
+ 2023-10-18 23:32:55,433 ----------------------------------------------------------------------------------------------------
93
+ 2023-10-18 23:33:01,085 epoch 2 - iter 361/3617 - loss 0.18186471 - time (sec): 5.65 - samples/sec: 6671.55 - lr: 0.000049 - momentum: 0.000000
94
+ 2023-10-18 23:33:06,832 epoch 2 - iter 722/3617 - loss 0.17555892 - time (sec): 11.40 - samples/sec: 6699.64 - lr: 0.000049 - momentum: 0.000000
95
+ 2023-10-18 23:33:12,500 epoch 2 - iter 1083/3617 - loss 0.18267083 - time (sec): 17.07 - samples/sec: 6654.16 - lr: 0.000048 - momentum: 0.000000
96
+ 2023-10-18 23:33:18,133 epoch 2 - iter 1444/3617 - loss 0.17984379 - time (sec): 22.70 - samples/sec: 6659.33 - lr: 0.000048 - momentum: 0.000000
97
+ 2023-10-18 23:33:23,799 epoch 2 - iter 1805/3617 - loss 0.17830915 - time (sec): 28.37 - samples/sec: 6679.06 - lr: 0.000047 - momentum: 0.000000
98
+ 2023-10-18 23:33:29,543 epoch 2 - iter 2166/3617 - loss 0.17452639 - time (sec): 34.11 - samples/sec: 6706.68 - lr: 0.000047 - momentum: 0.000000
99
+ 2023-10-18 23:33:35,235 epoch 2 - iter 2527/3617 - loss 0.17431735 - time (sec): 39.80 - samples/sec: 6692.15 - lr: 0.000046 - momentum: 0.000000
100
+ 2023-10-18 23:33:40,836 epoch 2 - iter 2888/3617 - loss 0.17180705 - time (sec): 45.40 - samples/sec: 6680.76 - lr: 0.000046 - momentum: 0.000000
101
+ 2023-10-18 23:33:46,319 epoch 2 - iter 3249/3617 - loss 0.17124739 - time (sec): 50.89 - samples/sec: 6708.37 - lr: 0.000045 - momentum: 0.000000
102
+ 2023-10-18 23:33:52,038 epoch 2 - iter 3610/3617 - loss 0.16926488 - time (sec): 56.60 - samples/sec: 6699.72 - lr: 0.000044 - momentum: 0.000000
103
+ 2023-10-18 23:33:52,139 ----------------------------------------------------------------------------------------------------
104
+ 2023-10-18 23:33:52,139 EPOCH 2 done: loss 0.1692 - lr: 0.000044
105
+ 2023-10-18 23:33:55,914 DEV : loss 0.16764183342456818 - f1-score (micro avg) 0.3918
106
+ 2023-10-18 23:33:55,943 saving best model
107
+ 2023-10-18 23:33:55,982 ----------------------------------------------------------------------------------------------------
108
+ 2023-10-18 23:34:01,814 epoch 3 - iter 361/3617 - loss 0.16873540 - time (sec): 5.83 - samples/sec: 6497.65 - lr: 0.000044 - momentum: 0.000000
109
+ 2023-10-18 23:34:07,470 epoch 3 - iter 722/3617 - loss 0.15570685 - time (sec): 11.49 - samples/sec: 6513.90 - lr: 0.000043 - momentum: 0.000000
110
+ 2023-10-18 23:34:13,134 epoch 3 - iter 1083/3617 - loss 0.15229953 - time (sec): 17.15 - samples/sec: 6685.33 - lr: 0.000043 - momentum: 0.000000
111
+ 2023-10-18 23:34:18,815 epoch 3 - iter 1444/3617 - loss 0.15006958 - time (sec): 22.83 - samples/sec: 6651.49 - lr: 0.000042 - momentum: 0.000000
112
+ 2023-10-18 23:34:24,508 epoch 3 - iter 1805/3617 - loss 0.14545533 - time (sec): 28.53 - samples/sec: 6701.89 - lr: 0.000042 - momentum: 0.000000
113
+ 2023-10-18 23:34:29,907 epoch 3 - iter 2166/3617 - loss 0.14270393 - time (sec): 33.92 - samples/sec: 6781.32 - lr: 0.000041 - momentum: 0.000000
114
+ 2023-10-18 23:34:35,612 epoch 3 - iter 2527/3617 - loss 0.14114846 - time (sec): 39.63 - samples/sec: 6742.66 - lr: 0.000041 - momentum: 0.000000
115
+ 2023-10-18 23:34:41,129 epoch 3 - iter 2888/3617 - loss 0.14188453 - time (sec): 45.15 - samples/sec: 6746.12 - lr: 0.000040 - momentum: 0.000000
116
+ 2023-10-18 23:34:46,190 epoch 3 - iter 3249/3617 - loss 0.14157282 - time (sec): 50.21 - samples/sec: 6815.93 - lr: 0.000039 - momentum: 0.000000
117
+ 2023-10-18 23:34:51,682 epoch 3 - iter 3610/3617 - loss 0.14134114 - time (sec): 55.70 - samples/sec: 6809.87 - lr: 0.000039 - momentum: 0.000000
118
+ 2023-10-18 23:34:51,787 ----------------------------------------------------------------------------------------------------
119
+ 2023-10-18 23:34:51,787 EPOCH 3 done: loss 0.1415 - lr: 0.000039
120
+ 2023-10-18 23:34:55,000 DEV : loss 0.16292423009872437 - f1-score (micro avg) 0.4771
121
+ 2023-10-18 23:34:55,028 saving best model
122
+ 2023-10-18 23:34:55,069 ----------------------------------------------------------------------------------------------------
123
+ 2023-10-18 23:35:00,739 epoch 4 - iter 361/3617 - loss 0.13410168 - time (sec): 5.67 - samples/sec: 6389.88 - lr: 0.000038 - momentum: 0.000000
124
+ 2023-10-18 23:35:06,521 epoch 4 - iter 722/3617 - loss 0.13650069 - time (sec): 11.45 - samples/sec: 6598.29 - lr: 0.000038 - momentum: 0.000000
125
+ 2023-10-18 23:35:12,008 epoch 4 - iter 1083/3617 - loss 0.13257245 - time (sec): 16.94 - samples/sec: 6715.49 - lr: 0.000037 - momentum: 0.000000
126
+ 2023-10-18 23:35:17,740 epoch 4 - iter 1444/3617 - loss 0.13291513 - time (sec): 22.67 - samples/sec: 6642.34 - lr: 0.000037 - momentum: 0.000000
127
+ 2023-10-18 23:35:23,361 epoch 4 - iter 1805/3617 - loss 0.13001197 - time (sec): 28.29 - samples/sec: 6659.44 - lr: 0.000036 - momentum: 0.000000
128
+ 2023-10-18 23:35:29,139 epoch 4 - iter 2166/3617 - loss 0.12690191 - time (sec): 34.07 - samples/sec: 6663.86 - lr: 0.000036 - momentum: 0.000000
129
+ 2023-10-18 23:35:34,906 epoch 4 - iter 2527/3617 - loss 0.12813304 - time (sec): 39.84 - samples/sec: 6686.69 - lr: 0.000035 - momentum: 0.000000
130
+ 2023-10-18 23:35:40,607 epoch 4 - iter 2888/3617 - loss 0.12911785 - time (sec): 45.54 - samples/sec: 6658.28 - lr: 0.000034 - momentum: 0.000000
131
+ 2023-10-18 23:35:46,281 epoch 4 - iter 3249/3617 - loss 0.12698770 - time (sec): 51.21 - samples/sec: 6679.26 - lr: 0.000034 - momentum: 0.000000
132
+ 2023-10-18 23:35:52,052 epoch 4 - iter 3610/3617 - loss 0.12601547 - time (sec): 56.98 - samples/sec: 6651.46 - lr: 0.000033 - momentum: 0.000000
133
+ 2023-10-18 23:35:52,170 ----------------------------------------------------------------------------------------------------
134
+ 2023-10-18 23:35:52,170 EPOCH 4 done: loss 0.1259 - lr: 0.000033
135
+ 2023-10-18 23:35:56,007 DEV : loss 0.17158399522304535 - f1-score (micro avg) 0.4922
136
+ 2023-10-18 23:35:56,035 saving best model
137
+ 2023-10-18 23:35:56,074 ----------------------------------------------------------------------------------------------------
138
+ 2023-10-18 23:36:01,842 epoch 5 - iter 361/3617 - loss 0.10767703 - time (sec): 5.77 - samples/sec: 6797.38 - lr: 0.000033 - momentum: 0.000000
139
+ 2023-10-18 23:36:07,648 epoch 5 - iter 722/3617 - loss 0.11538990 - time (sec): 11.57 - samples/sec: 6711.32 - lr: 0.000032 - momentum: 0.000000
140
+ 2023-10-18 23:36:13,368 epoch 5 - iter 1083/3617 - loss 0.10959815 - time (sec): 17.29 - samples/sec: 6725.29 - lr: 0.000032 - momentum: 0.000000
141
+ 2023-10-18 23:36:19,117 epoch 5 - iter 1444/3617 - loss 0.10966842 - time (sec): 23.04 - samples/sec: 6697.82 - lr: 0.000031 - momentum: 0.000000
142
+ 2023-10-18 23:36:24,772 epoch 5 - iter 1805/3617 - loss 0.11133696 - time (sec): 28.70 - samples/sec: 6661.49 - lr: 0.000031 - momentum: 0.000000
143
+ 2023-10-18 23:36:30,368 epoch 5 - iter 2166/3617 - loss 0.11272880 - time (sec): 34.29 - samples/sec: 6649.42 - lr: 0.000030 - momentum: 0.000000
144
+ 2023-10-18 23:36:36,026 epoch 5 - iter 2527/3617 - loss 0.11283559 - time (sec): 39.95 - samples/sec: 6669.94 - lr: 0.000029 - momentum: 0.000000
145
+ 2023-10-18 23:36:41,522 epoch 5 - iter 2888/3617 - loss 0.11205737 - time (sec): 45.45 - samples/sec: 6696.03 - lr: 0.000029 - momentum: 0.000000
146
+ 2023-10-18 23:36:46,562 epoch 5 - iter 3249/3617 - loss 0.11215470 - time (sec): 50.49 - samples/sec: 6768.50 - lr: 0.000028 - momentum: 0.000000
147
+ 2023-10-18 23:36:52,200 epoch 5 - iter 3610/3617 - loss 0.11228934 - time (sec): 56.13 - samples/sec: 6756.66 - lr: 0.000028 - momentum: 0.000000
148
+ 2023-10-18 23:36:52,312 ----------------------------------------------------------------------------------------------------
149
+ 2023-10-18 23:36:52,312 EPOCH 5 done: loss 0.1122 - lr: 0.000028
150
+ 2023-10-18 23:36:55,548 DEV : loss 0.18035954236984253 - f1-score (micro avg) 0.5056
151
+ 2023-10-18 23:36:55,576 saving best model
152
+ 2023-10-18 23:36:55,615 ----------------------------------------------------------------------------------------------------
153
+ 2023-10-18 23:37:01,448 epoch 6 - iter 361/3617 - loss 0.10582480 - time (sec): 5.83 - samples/sec: 6762.34 - lr: 0.000027 - momentum: 0.000000
154
+ 2023-10-18 23:37:07,097 epoch 6 - iter 722/3617 - loss 0.10344347 - time (sec): 11.48 - samples/sec: 6653.64 - lr: 0.000027 - momentum: 0.000000
155
+ 2023-10-18 23:37:12,714 epoch 6 - iter 1083/3617 - loss 0.10584413 - time (sec): 17.10 - samples/sec: 6554.94 - lr: 0.000026 - momentum: 0.000000
156
+ 2023-10-18 23:37:18,518 epoch 6 - iter 1444/3617 - loss 0.10479511 - time (sec): 22.90 - samples/sec: 6591.41 - lr: 0.000026 - momentum: 0.000000
157
+ 2023-10-18 23:37:24,193 epoch 6 - iter 1805/3617 - loss 0.10413115 - time (sec): 28.58 - samples/sec: 6591.53 - lr: 0.000025 - momentum: 0.000000
158
+ 2023-10-18 23:37:29,887 epoch 6 - iter 2166/3617 - loss 0.10251088 - time (sec): 34.27 - samples/sec: 6634.11 - lr: 0.000024 - momentum: 0.000000
159
+ 2023-10-18 23:37:35,572 epoch 6 - iter 2527/3617 - loss 0.09916624 - time (sec): 39.96 - samples/sec: 6621.63 - lr: 0.000024 - momentum: 0.000000
160
+ 2023-10-18 23:37:41,386 epoch 6 - iter 2888/3617 - loss 0.09901356 - time (sec): 45.77 - samples/sec: 6621.04 - lr: 0.000023 - momentum: 0.000000
161
+ 2023-10-18 23:37:47,445 epoch 6 - iter 3249/3617 - loss 0.10012522 - time (sec): 51.83 - samples/sec: 6597.53 - lr: 0.000023 - momentum: 0.000000
162
+ 2023-10-18 23:37:53,206 epoch 6 - iter 3610/3617 - loss 0.10172452 - time (sec): 57.59 - samples/sec: 6585.83 - lr: 0.000022 - momentum: 0.000000
163
+ 2023-10-18 23:37:53,310 ----------------------------------------------------------------------------------------------------
164
+ 2023-10-18 23:37:53,310 EPOCH 6 done: loss 0.1017 - lr: 0.000022
165
+ 2023-10-18 23:37:56,522 DEV : loss 0.19675783812999725 - f1-score (micro avg) 0.52
166
+ 2023-10-18 23:37:56,550 saving best model
167
+ 2023-10-18 23:37:56,582 ----------------------------------------------------------------------------------------------------
168
+ 2023-10-18 23:38:02,251 epoch 7 - iter 361/3617 - loss 0.09799474 - time (sec): 5.67 - samples/sec: 6829.25 - lr: 0.000022 - momentum: 0.000000
169
+ 2023-10-18 23:38:07,989 epoch 7 - iter 722/3617 - loss 0.09523192 - time (sec): 11.41 - samples/sec: 6830.30 - lr: 0.000021 - momentum: 0.000000
170
+ 2023-10-18 23:38:13,729 epoch 7 - iter 1083/3617 - loss 0.09715759 - time (sec): 17.15 - samples/sec: 6717.61 - lr: 0.000021 - momentum: 0.000000
171
+ 2023-10-18 23:38:19,390 epoch 7 - iter 1444/3617 - loss 0.09700347 - time (sec): 22.81 - samples/sec: 6691.41 - lr: 0.000020 - momentum: 0.000000
172
+ 2023-10-18 23:38:25,140 epoch 7 - iter 1805/3617 - loss 0.09588648 - time (sec): 28.56 - samples/sec: 6685.91 - lr: 0.000019 - momentum: 0.000000
173
+ 2023-10-18 23:38:30,875 epoch 7 - iter 2166/3617 - loss 0.09457137 - time (sec): 34.29 - samples/sec: 6695.36 - lr: 0.000019 - momentum: 0.000000
174
+ 2023-10-18 23:38:36,556 epoch 7 - iter 2527/3617 - loss 0.09554146 - time (sec): 39.97 - samples/sec: 6686.03 - lr: 0.000018 - momentum: 0.000000
175
+ 2023-10-18 23:38:42,335 epoch 7 - iter 2888/3617 - loss 0.09478297 - time (sec): 45.75 - samples/sec: 6661.58 - lr: 0.000018 - momentum: 0.000000
176
+ 2023-10-18 23:38:47,875 epoch 7 - iter 3249/3617 - loss 0.09584686 - time (sec): 51.29 - samples/sec: 6673.39 - lr: 0.000017 - momentum: 0.000000
177
+ 2023-10-18 23:38:53,473 epoch 7 - iter 3610/3617 - loss 0.09639512 - time (sec): 56.89 - samples/sec: 6668.95 - lr: 0.000017 - momentum: 0.000000
178
+ 2023-10-18 23:38:53,574 ----------------------------------------------------------------------------------------------------
179
+ 2023-10-18 23:38:53,574 EPOCH 7 done: loss 0.0964 - lr: 0.000017
180
+ 2023-10-18 23:38:57,413 DEV : loss 0.20494325459003448 - f1-score (micro avg) 0.5123
181
+ 2023-10-18 23:38:57,441 ----------------------------------------------------------------------------------------------------
182
+ 2023-10-18 23:39:03,091 epoch 8 - iter 361/3617 - loss 0.08316593 - time (sec): 5.65 - samples/sec: 6661.87 - lr: 0.000016 - momentum: 0.000000
183
+ 2023-10-18 23:39:08,961 epoch 8 - iter 722/3617 - loss 0.08133314 - time (sec): 11.52 - samples/sec: 6552.02 - lr: 0.000016 - momentum: 0.000000
184
+ 2023-10-18 23:39:14,650 epoch 8 - iter 1083/3617 - loss 0.08367591 - time (sec): 17.21 - samples/sec: 6644.50 - lr: 0.000015 - momentum: 0.000000
185
+ 2023-10-18 23:39:20,408 epoch 8 - iter 1444/3617 - loss 0.08496144 - time (sec): 22.97 - samples/sec: 6612.09 - lr: 0.000014 - momentum: 0.000000
186
+ 2023-10-18 23:39:26,204 epoch 8 - iter 1805/3617 - loss 0.08735194 - time (sec): 28.76 - samples/sec: 6628.41 - lr: 0.000014 - momentum: 0.000000
187
+ 2023-10-18 23:39:31,922 epoch 8 - iter 2166/3617 - loss 0.09141252 - time (sec): 34.48 - samples/sec: 6648.80 - lr: 0.000013 - momentum: 0.000000
188
+ 2023-10-18 23:39:37,572 epoch 8 - iter 2527/3617 - loss 0.09151271 - time (sec): 40.13 - samples/sec: 6675.19 - lr: 0.000013 - momentum: 0.000000
189
+ 2023-10-18 23:39:43,240 epoch 8 - iter 2888/3617 - loss 0.09206089 - time (sec): 45.80 - samples/sec: 6672.07 - lr: 0.000012 - momentum: 0.000000
190
+ 2023-10-18 23:39:48,868 epoch 8 - iter 3249/3617 - loss 0.09102678 - time (sec): 51.43 - samples/sec: 6683.90 - lr: 0.000012 - momentum: 0.000000
191
+ 2023-10-18 23:39:54,734 epoch 8 - iter 3610/3617 - loss 0.08981402 - time (sec): 57.29 - samples/sec: 6623.49 - lr: 0.000011 - momentum: 0.000000
192
+ 2023-10-18 23:39:54,831 ----------------------------------------------------------------------------------------------------
193
+ 2023-10-18 23:39:54,831 EPOCH 8 done: loss 0.0898 - lr: 0.000011
194
+ 2023-10-18 23:39:58,030 DEV : loss 0.22380779683589935 - f1-score (micro avg) 0.5261
195
+ 2023-10-18 23:39:58,058 saving best model
196
+ 2023-10-18 23:39:58,091 ----------------------------------------------------------------------------------------------------
197
+ 2023-10-18 23:40:03,819 epoch 9 - iter 361/3617 - loss 0.07389080 - time (sec): 5.73 - samples/sec: 6734.41 - lr: 0.000011 - momentum: 0.000000
198
+ 2023-10-18 23:40:09,496 epoch 9 - iter 722/3617 - loss 0.08048499 - time (sec): 11.40 - samples/sec: 6728.63 - lr: 0.000010 - momentum: 0.000000
199
+ 2023-10-18 23:40:15,212 epoch 9 - iter 1083/3617 - loss 0.08163769 - time (sec): 17.12 - samples/sec: 6682.20 - lr: 0.000009 - momentum: 0.000000
200
+ 2023-10-18 23:40:21,028 epoch 9 - iter 1444/3617 - loss 0.08528220 - time (sec): 22.94 - samples/sec: 6708.43 - lr: 0.000009 - momentum: 0.000000
201
+ 2023-10-18 23:40:26,615 epoch 9 - iter 1805/3617 - loss 0.08612361 - time (sec): 28.52 - samples/sec: 6640.28 - lr: 0.000008 - momentum: 0.000000
202
+ 2023-10-18 23:40:32,305 epoch 9 - iter 2166/3617 - loss 0.08425714 - time (sec): 34.21 - samples/sec: 6642.71 - lr: 0.000008 - momentum: 0.000000
203
+ 2023-10-18 23:40:38,017 epoch 9 - iter 2527/3617 - loss 0.08439459 - time (sec): 39.93 - samples/sec: 6664.00 - lr: 0.000007 - momentum: 0.000000
204
+ 2023-10-18 23:40:43,797 epoch 9 - iter 2888/3617 - loss 0.08538617 - time (sec): 45.70 - samples/sec: 6657.30 - lr: 0.000007 - momentum: 0.000000
205
+ 2023-10-18 23:40:49,295 epoch 9 - iter 3249/3617 - loss 0.08512273 - time (sec): 51.20 - samples/sec: 6678.10 - lr: 0.000006 - momentum: 0.000000
206
+ 2023-10-18 23:40:54,945 epoch 9 - iter 3610/3617 - loss 0.08581878 - time (sec): 56.85 - samples/sec: 6673.93 - lr: 0.000006 - momentum: 0.000000
207
+ 2023-10-18 23:40:55,051 ----------------------------------------------------------------------------------------------------
208
+ 2023-10-18 23:40:55,052 EPOCH 9 done: loss 0.0858 - lr: 0.000006
209
+ 2023-10-18 23:40:58,910 DEV : loss 0.23317401111125946 - f1-score (micro avg) 0.5297
210
+ 2023-10-18 23:40:58,938 saving best model
211
+ 2023-10-18 23:40:58,977 ----------------------------------------------------------------------------------------------------
212
+ 2023-10-18 23:41:05,019 epoch 10 - iter 361/3617 - loss 0.08574863 - time (sec): 6.04 - samples/sec: 6107.84 - lr: 0.000005 - momentum: 0.000000
213
+ 2023-10-18 23:41:10,700 epoch 10 - iter 722/3617 - loss 0.08392903 - time (sec): 11.72 - samples/sec: 6349.28 - lr: 0.000004 - momentum: 0.000000
214
+ 2023-10-18 23:41:16,371 epoch 10 - iter 1083/3617 - loss 0.07922998 - time (sec): 17.39 - samples/sec: 6427.48 - lr: 0.000004 - momentum: 0.000000
215
+ 2023-10-18 23:41:21,800 epoch 10 - iter 1444/3617 - loss 0.08164333 - time (sec): 22.82 - samples/sec: 6551.76 - lr: 0.000003 - momentum: 0.000000
216
+ 2023-10-18 23:41:27,491 epoch 10 - iter 1805/3617 - loss 0.07922620 - time (sec): 28.51 - samples/sec: 6589.30 - lr: 0.000003 - momentum: 0.000000
217
+ 2023-10-18 23:41:33,314 epoch 10 - iter 2166/3617 - loss 0.08341687 - time (sec): 34.34 - samples/sec: 6620.00 - lr: 0.000002 - momentum: 0.000000
218
+ 2023-10-18 23:41:38,972 epoch 10 - iter 2527/3617 - loss 0.08209356 - time (sec): 39.99 - samples/sec: 6620.51 - lr: 0.000002 - momentum: 0.000000
219
+ 2023-10-18 23:41:44,642 epoch 10 - iter 2888/3617 - loss 0.08287713 - time (sec): 45.66 - samples/sec: 6626.57 - lr: 0.000001 - momentum: 0.000000
220
+ 2023-10-18 23:41:50,388 epoch 10 - iter 3249/3617 - loss 0.08215472 - time (sec): 51.41 - samples/sec: 6663.09 - lr: 0.000001 - momentum: 0.000000
221
+ 2023-10-18 23:41:55,996 epoch 10 - iter 3610/3617 - loss 0.08275437 - time (sec): 57.02 - samples/sec: 6654.92 - lr: 0.000000 - momentum: 0.000000
222
+ 2023-10-18 23:41:56,096 ----------------------------------------------------------------------------------------------------
223
+ 2023-10-18 23:41:56,097 EPOCH 10 done: loss 0.0830 - lr: 0.000000
224
+ 2023-10-18 23:41:59,297 DEV : loss 0.2345695048570633 - f1-score (micro avg) 0.5275
225
+ 2023-10-18 23:41:59,357 ----------------------------------------------------------------------------------------------------
226
+ 2023-10-18 23:41:59,357 Loading model from best epoch ...
227
+ 2023-10-18 23:41:59,438 SequenceTagger predicts: Dictionary with 13 tags: O, S-loc, B-loc, E-loc, I-loc, S-pers, B-pers, E-pers, I-pers, S-org, B-org, E-org, I-org
228
+ 2023-10-18 23:42:03,543
229
+ Results:
230
+ - F-score (micro) 0.5325
231
+ - F-score (macro) 0.3573
232
+ - Accuracy 0.375
233
+
234
+ By class:
235
+ precision recall f1-score support
236
+
237
+ loc 0.5215 0.6971 0.5967 591
238
+ pers 0.4273 0.5350 0.4751 357
239
+ org 0.0000 0.0000 0.0000 79
240
+
241
+ micro avg 0.4871 0.5871 0.5325 1027
242
+ macro avg 0.3163 0.4107 0.3573 1027
243
+ weighted avg 0.4486 0.5871 0.5085 1027
244
+
245
+ 2023-10-18 23:42:03,543 ----------------------------------------------------------------------------------------------------